skip to navigation
skip to content

pussycache 1.3

Cache Backend system for python objects

Latest Version: 1.4

Python pussy cache

Python pussy cache is a Cache system for python objects.

Cache backends can be in-memory or redis/ You can even use the django
cache framework with Python-pussy-cache.

Given a one of the cache backends and any Python class you have
define, Python-pussy-cache will cache the results of methods you have
define and will also manage the cache invalidation by timestamp or
with methods you have defined.

[![Build Status](](

Here is an example to make the thing clearer

import time
from pussycache.proxy import BaseProxy
from pussycache.cache import BaseCacheBackend
# Here is a simple class where some methods need to be cached

class MyClass(object):

def a_long_task(self, delta):
return delta

def forget_about_time(self):
return None

# We set an in memory cache backend with a TTL of 30 seconds.

cache = BaseCacheBackend(30)

cache_proxy = BaseProxy(MyClass(), cache=cache,
invalidate_methods={"forget_about_time": ["a_long_task"]})

cachedinstance = cache_proxy
# your cachedinstance is now
# ready to use. It will just work like a regular MyClass object

print cachedinstance.a_long_task(10)
# 10 seconds later
# if we call the same method a second time:
print cachedinstance.a_long_task(10)
# result is returned immediatly because it's in the cache
# but
print cachedinstance.a_long_task(3)
# with different parameters, the result is not cached yet.
# if you want to invalidate the cache for this method:
print cachedinstance.forget_about_time()
print cachedinstance.a_long_task(10)
# 10 seconds later

Of course, If you need direct access to the cache backend, you can
call it directly. Let's say you need to invalidate ALL the cache :



The same apply if you need to call directly the cache:

cachedinstance.\_cache.set("mykey", "my value", 10)
"my value"
#and 10 seconds later:
#the value is gone from the cache


To run test, just install tox with ``pip install tox`` and run

File Type Py Version Uploaded on Size (md5) Source 2014-02-05 11KB